INDEX

Explanations

comprehensive/fundamental concepts

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

irc

0.78

sister

0.74

 sister

0.73

 cousin

0.71

ibling

0.71

elta

0.71

 bakt

0.70

}^{

0.69

 surrog

0.68

 sibling

0.68

POSITIVE LOGITS

 suficientes

1.50

 COMPLETE

1.50

 đủ

1.44

COMPLETE

1.43

彻底

1.38

 profound

1.37

品质

1.36

 profoundly

1.34

 noticeably

1.33

ysław

1.33

Activations Density 0.225%